OpenAI Noam Brown says single-number benchmarks no longer make sense for modern AI models Once models can use CoT and extra inference compute, their performance depends heavily on how much reasoning time they are given "it made sense for GPT-2/3/4, but not for reasoning models"

@haider1 @polynoamial hook me up with that hoodie big dawg

@haider1 Yeah, Compute and Inference Engineering are like .com of 2026